Author Topic: EMI Detection, difference when using inductors, antennas, and proper EMI probes?  (Read 3147 times)

0 Members and 1 Guest are viewing this topic.

Offline rwgast_lowlevellogicdesinTopic starter

  • Frequent Contributor
  • **
  • Posts: 591
  • Country: us
    • LowLevel-LogicDesign
This isnt all about RF per say, but I figured anyone in this section would probably understand my question I am getting a little hung up on a small project....

A while back someone sent this product to Dave in a mailbag

Elecktrosluch3 http://www.synthtopia.com/content/2015/07/31/elektrosluch-3-like-a-mic-for-electromagnetic-fields/
Schematic https://dl.dropboxusercontent.com/u/27653605/elektrosluch_scheme.pdf

From the schematic you can see that its just two 22mH inductors ran through an opamp... pretty damn simple. Now obviously this thing will only be viable with fields from 20hz-20Khz or any harmonics produced in the human hearing range, If one were to scrap the audio jack and connect it to a scope it would be a lot more useful for other frequencies... what range im not sure. To me this is about the same thing as connecting an old "telephone pickup" to an amp or scope.

Then we have this, the amazing all band all at once semi direction reciver
http://www.techlib.com/electronics/allband.htm

Now I am no guru here but you have an antenna connected to a crystal or low drop shottky in this case and then amplified before finally sent to an audio amp like the LM386. With no mixer how would you be able to detect any frequencies out of the hearing range? The human ear cant hear 100mhz..? So what am I missing that makes this work?

And finally you have professional EMI probe kits that can also be DIYed, well what if we used one of those as the "sensor" in one of the devices above?

I would like to make one or both of the circuits above so I can listen to EMI/RF when trouble shooting and also add a BNC jack to pplug in to a scope for better and visual analyzation but I am not really sure what the difference is between using inductors or antennas #1, and im also having a hard time grasping the concept of hearing frequencies about 17Khz without a mixer to do broadband down conversion to audio. Lastly what do proper probes offer over the first two methods and could the be combined with them to make quick EMI/RF interference problems readly apparent without a scope but with audio?

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21606
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
From the schematic you can see that its just two 22mH inductors ran through an opamp... pretty damn simple. Now obviously this thing will only be viable with fields from 20hz-20Khz or any harmonics produced in the human hearing range, If one were to scrap the audio jack and connect it to a scope it would be a lot more useful for other frequencies... what range im not sure. To me this is about the same thing as connecting an old "telephone pickup" to an amp or scope.

For low frequencies, yes, a large inductance will do fine, and for the same reasons and purposes as a "telephone pickup".

Quote
Then we have this, the amazing all band all at once semi direction reciver
http://www.techlib.com/electronics/allband.htm

Now I am no guru here but you have an antenna connected to a crystal or low drop shottky in this case and then amplified before finally sent to an audio amp like the LM386. With no mixer how would you be able to detect any frequencies out of the hearing range? The human ear cant hear 100mhz..? So what am I missing that makes this work?

It detects the total envelope and you hear that.  If the loudest channel isn't the one you wanted to hear... too bad.

What's more, if there are multiple nearby stations with similar amplitudes (give or take multipath, perhaps?), you'll get all of them mixed together!

And of FM, you'll hear nothing*.

*Multipath tends to "slope detect" a good bit of FM, though it's almost always distorted, and the frequency response is wrong (due to preemphasis).  This is largely why radio can often be heard through cheap amplified speakers and whatnot.  Or at least the high frequency content that's full of cymbals and consonants, easily discerned despite the distortion.

Quote
And finally you have professional EMI probe kits that can also be DIYed, well what if we used one of those as the "sensor" in one of the devices above?

I would like to make one or both of the circuits above so I can listen to EMI/RF when trouble shooting and also add a BNC jack to pplug in to a scope for better and visual analyzation but I am not really sure what the difference is between using inductors or antennas #1, and im also having a hard time grasping the concept of hearing frequencies about 17Khz without a mixer to do broadband down conversion to audio. Lastly what do proper probes offer over the first two methods and could the be combined with them to make quick EMI/RF interference problems readly apparent without a scope but with audio?

For EMI/RF, you don't want big inductors, you want small ones.  I made a one of this, a long time ago:



it's unshielded and unbalanced, so it also picks up some E-field, and it's a simple loop so the sensitivity has a dipole shape.  The signal level is quite low so it's best paired with a wideband preamp.  If you add an integrator as well, you get a qualitative measure of the current flows that are creating your signal.

The other kind you see often, is a coil shielded by a slotted tube; this still has a dipole response, but the one lobe is physically unavailable (you can't point the cable end at the circuit, because the tube is in the way!), and the slit skews the lobe, so the peak response is tilted towards the slitted side.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline rwgast_lowlevellogicdesinTopic starter

  • Frequent Contributor
  • **
  • Posts: 591
  • Country: us
    • LowLevel-LogicDesign
Hmmm thanks for the info, I was pretty sure the inductor method was for LF only and works somewhat like a loaded antenna?

As far as your probe I have mad a few, your design looks much different, mine are just coax looped and the center conductor is soldiered to the sheild at the loops junction. Would you mind posting the details of your probe? I see the transistors there I am guessing that those are for the broadband pre, what model are they.

I would like to build a loop like that with a BandWidth of 25mhz. Also how do the voltage reading on the scope directly compare to if you were to probe the same spot and check voltages on the scope with acuall probes.

Secondly Im thinking od doing something a bit "dangerous" but in my case I think its ok. I would like to lift the scopes ground when using EMi probes. I have noticed that if I plug a noisey switcher in to the same circuit in to the power strip my scope is on, all the garbage noise going back in to the power lines couples in to my scope. Im looking to get a noise free display when doing EMI/RF probing.

Also if you float the scope and measure two small DC voltage particularly on a circuit with low voltage differential ADC connections, and the DUT is battery powered would working with a floated ground be ok in this situation and help make accurate measurements using the two probe subtract method as a differential probe.

These operations would be the only be used for EMI and on small differential ADC/DAC measurements from another isolated device. Lastly if one goes this path how about cheater plug lifting vs isolation transformer?

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21606
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
The transistors are for my... look at my avatar. :P

The probe is what you see here; nothing is hidden under the heatshrink, only some ugly soldered connections from wire to coax.

The bandwidth of this loop is probably several 100 MHz, at which point the E vs. M discrimination will be useless, and the electrical length of the loop itself becomes important.  The loss of the RG-174 will also have a small impact.

Lifting scope ground will do absolutely nothing above ~kHz.  Common mode filtering at the probe can help (to the tune of 3-10dB), but maintaining consistent grounding and shielding is the only sure approach (to the tune of 20-100+ dB).

The capacitance (or equivalent) to an E/M probe should be vanishingly small anyway, so that the common mode current is much smaller than the signal (unless that's what you're measuring, as is the case for an E probe).

Conducted noise from a DUT is dealt with by powering it through a shielded and grounded LISN.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline rwgast_lowlevellogicdesinTopic starter

  • Frequent Contributor
  • **
  • Posts: 591
  • Country: us
    • LowLevel-LogicDesign
So floating the scope, will work well for audio applications? This is where I get the problems especially in the FFT, if I plug a dirty wal wart in to the same house circuit my FFT goues nuts!

what is LISN?

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21606
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
LISN = Line Impedance Stabilization Network. Filters the incoming AC line, and couples the load to a consistent (50 ohms) impedance for testing.  The 50 ohms can be a termination resistor, a spectrum analyzer (or anything else to measure the conducted emissions), or an RF source (for conducted susceptibility testing).

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf